检索结果-内蒙古大学图书馆

Experimental Verification of a Mathematical Model for a Heat Store

Russian Engineering Research 2021年第8期41卷 742-744页

作者： Sorokin, A.E. Moscow Aviation Institute Moscow Russia

Abstract: A method is proposed for the analysis of experimental data so as to verify the mathematical model of a heat store with a melting working medium. Topics considered include the verification of the experimental data;elimination of discrepant values;approximation of the experimental data;and verification of the approximate formula for the experimental data. © 2021, Allerton Press, Inc.

关键词： approximation algorithms

来源：评论

学校读者我要写书评

暂无评论

Influence Maximization with Fairness Cost on Groups in Online Social Networks 17th

Influence Maximization with Fairness Cost on Groups in On...

引用

17th Asian Conference on Intelligent Information and Database Systems, ACIIDS 2025

作者： Nguyen, Hue T. Pham, Bac D. Ha, Dung T. K. Giang, Nguyen Long Pham, Canh V. Hanoi Viet Nam Faculty of Information Technology Hanoi Architecture University Hanoi Viet Nam People’s Security Academy Hanoi Viet Nam ORLab Phenikaa University Hanoi12116 Viet Nam

ISBN: (纸本)9789819660049

Motivated by applications in social influence and viral marketing, this work introduces the Influence Maximization with Cost Fairness on Groups (IMFC) problem in social networks. Given a social network, a total budget, and a set of user groups, the goal is to identify a subset of users that maximizes influence spread while ensuring that the total cost within each group does not exceed its allocated budget. However, existing algorithms for influence maximization problems may not adapt to the problem or do not give any theoretical guarantees. To address the challenge, in this work we propose an efficient and scalable approximation algorithm that returns an approximation ratio of αα+2-ϵ and takes O((m+log(nϵ))nϵ2(nlogn+log(1δ))) time complexity, where α is the fairness ratio of cost, ϵ,δ are constants, n is the number of uses, and m is the number of links. Beyond theoretical guarantees, we conduct extensive experiments on real-world social networks. Our results demonstrate that the proposed algorithm not only achieves better influence spread but also runs faster than state-of-the-art methods. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： approximation algorithms

来源：评论

学校读者我要写书评

暂无评论

Quaternionic Minkowski Reduction and Its Application in Dual-Polarized MIMO Systems

Quaternionic Minkowski Reduction and Its Application in Dual...

引用

International ITG Conference on Source and Channel Coding (SCC)

作者： Sebastian Stern Cong Ling Cornelia Ott Robert F.H. Fischer Institute of Communications Engineering Ulm University Germany Department of Electrical and Electronic Engineering Imperial College London United Kingdom Institute of Communications and Navigation German Aerospace Center (DLR) Germany

ISBN: (数字)9798331522896

ISBN: (纸本)9798331522902

The task of finding the shortest vectors of a lattice that form a basis thereof is known as Minkowski reduction. In this work, Minkowski reduction is adapted to discrete subgroups of quaternionic spaces defined over the set of Hurwitz integers which constitutes a Euclidean subring of the (Hamilton) quaternions. Based on a quaternionic variant of Minkowski's greatest-common-divisor condition for suitable lattice vectors, a lattice-basis-reduction algorithm is derived. To that end, a basisupdate approach is presented which takes the non-commutativity of quaternionic multiplication into account. The reduction algorithm is finally applied in multi-user multiple-input/multipleoutput (MIMO) systems with dual-polarized antennas, where the quaternionic interference is handled by lattice-reduction-aided equalization. It is shown that quaternionic Minkowski reduction achieves significantly better results than state-of-the-art Lenstra-Lenstra-Lovász (LLL) reduction and that the successive minima of quaternionic lattices are approximated quite well.

关键词： Quaternions Lattices Interference approximation algorithms Vectors MIMO Encoding Antennas

来源：评论

学校读者我要写书评

暂无评论

Algorithm Design for Continual Learning in IoT Networks

Algorithm Design for Continual Learning in IoT Networks

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Shugang Hao Lingjie Duan Pillar of Engineering Systems and Design Singapore University of Technology and Design

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Continual learning (CL) is a new online learning technique over sequentially generated streaming data from different tasks, aiming to maintain a small forgetting loss on previously-learned tasks. Existing work focuses on reducing the forgetting loss under a given task sequence. However, if similar tasks continuously appear to the end time, the forgetting loss is still huge on prior distinct tasks. In practical IoT networks, an autonomous vehicle to sample data and learn different tasks can route and alter the order of task pattern at increased travelling cost. To our best knowledge, we are the first to study how to opportunistically route the testing object and alter the task sequence in CL. We formulate a new optimization problem and prove it NP-hard. We propose a polynomial-time algorithm to achieve approximation ratios of $\frac{3}{2}$ for underparameterized case and $\frac{3}{2} + {r^{1 - T}}$ for overparameterized case, respectively. Simulation results verify our algorithm’s close-to-optimum performance.

关键词： Continuing education Costs Simulation Signal processing algorithms Signal processing approximation algorithms Polynomials Speech processing Optimization Testing

来源：评论

学校读者我要写书评

暂无评论

Redefining Platform Data Mining Paradigms with Sequential Quadratic Programming Considering Optimization Models

Redefining Platform Data Mining Paradigms with Sequential Qu...

引用

Sentiment Analysis and Deep Learning (ICSADL), International Conference on

作者： Xiong Wenting Guangdong University of Science and Technology Dongguan Guangdong China

ISBN: (数字)9798331523923

ISBN: (纸本)9798331523930

Platform data mining is an important branch of data analysis. Traditional methods such as clustering have achieved satisfactory performance. To overcome the shortcomings of traditional algorithms in mathematical optimization, this study proposes a novel model based on the sequential quadratic programming algorithm. At the algorithm level, this study first analyses the characteristics of cloud platform data and its requirements for mining efficiency. At the same time, to solve these problems, this study proposed a new data analysis framework that combines sparse factor analysis and embedded database subspace detection. The designed model optimizes attribute dimension selection and dense region extraction to the greatest extent through distribution analysis and feature correlation evaluation. At the same time, the Bayesian network node expansion algorithm is used to model the association of discrete data, and then a cascade data generation method is designed based on this model. Finally, the Bayesian network parameters are optimized by the sequential quadratic programming algorithm, and the approximate value of the Hessian matrix is efficiently solved by the BFGS algorithm, thereby improving the accuracy of the data mining algorithm. The experimental part uses the cloud platform data-set as the target data-set and verifies the stability of the proposed algorithm.

关键词： Cloud computing Heuristic algorithms Computational modeling Clustering algorithms approximation algorithms Feature extraction Data models Data mining Quadratic programming Optimization

来源：评论

学校读者我要写书评

暂无评论

Embedding Graphs as Euclidean kNN-Graphs

arXiv

引用

arXiv 2025年

作者： Schibler, Thomas Suri, Subhash Xue, Jie University of California Santa BarbaraCA United States New York University Shanghai China

Let G = (V, E) be a directed graph on n vertices where each vertex has out-degree k. We say that G is kNN-realizable in d-dimensional Euclidean space if there exists a point set P = {p1, p2, . . ., pn} in Rd along with a one-to-one mapping : V → P such that for any u, v ∈ V , u is an out-neighbor of v in G if and only if (u) is one of the k nearest neighbors of (v);we call the map a kNN-realization of G in Rd. The kNN-realization problem, which aims to compute a kNN-realization of an input graph in Rd, is known to be NP-hard already for d = 2 and k = 1 [Eades and Whitesides, Theoretical Computer Science, 1996], and to the best of our knowledge has not been studied in dimension d = 1. The main results of this paper are the following: For any fixed dimension d ≥ 2, we can efficiently compute an embedding realizing at least a 1 − Ε fraction of G’s edges, or conclude that G is not kNN-realizable in Rd. For d = 1, we can decide in O(kn) time whether G is kNN-realizable and, if so, compute a realization in O(n2.5poly(log n)) time. © 2025, CC BY.

关键词： approximation algorithms

来源：评论

学校读者我要写书评

暂无评论

Fair Set Cover 25

Fair Set Cover

引用

Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.1

作者： Mohsen Dehghankar Rahul Raychaudhury Stavros Sintos Abolfazl Asudeh University of Illinois Chicago Chicago IL USA Duke University Durham NC USA University of Illinois Chicago Chicago USA

ISBN: (纸本)9798400712456

The potential harms of algorithmic decisions have ignited algorithmic fairness as a central topic in computer science. One of the fundamental problems in computer science is Set Cover, which has numerous applications with societal impacts, such as assembling a small team of individuals that collectively satisfy a range of expertise requirements. However, despite its broad application spectrum and significant potential impact, set cover has yet to be studied through the lens of ***, in this paper, we introduce Fair Set Cover, which aims not only to cover with a minimum-size set but also to satisfy demographic parity in its selection of sets. To this end, we develop multiple versions of fair set cover, study their hardness, and devise efficient approximation algorithms for each variant. Notably, under certain assumptions, our algorithms always guarantee zero-unfairness, with only a small increase in the approximation ratio compared to regular set cover. Furthermore, our experiments on various data sets and across different settings confirm the negligible price of fairness, as (a) the output size increases only slightly (if any) and (b) the time to compute the output does not significantly increase.

关键词： approximation algorithms

来源：评论

学校读者我要写书评

暂无评论

A Geometry-Based Node Activation Method for Relative Localization

A Geometry-Based Node Activation Method for Relative Localiz...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Licheng Wang Yi Li Hanying Zhao Yuan Shen Department of Electronic Engineering Beijing National Research Center for Information Science and Technology Tsinghua University Beijing China School of Electronic Engineering and Computer Science KTH Royal Institute of Technology Stockholm Sweden Shanghai Artificial Intelligence Laboratory Shanghai China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

In multi-agent systems, the hybrid active-silent relative localization framework is widely employed, where only active nodes transmit signals. The selection of active nodes, known as node activation, significantly impacts the positioning accuracy. This paper investigates the node activation in anchor-free localization systems. First, the constrained Cramér-Rao lower bound (CRLB) is derived to evaluate the localization error. Then the combinatorial optimization problem on node activation is presented and approximately solved using the difference of convex programming (DCP) method. Moreover, to reduce computational complexity, we propose a geometry-based greedy iterative (GBGI) algorithm which leverages a geometry metric to evaluate and iteratively refine the selection of active nodes. Finally, simulation results demonstrate the performance of proposed algorithms. Especially the GBGI algorithm closely approaches the optimal solution.

关键词： Location awareness Measurement Geometry Accuracy Simulation Signal processing algorithms Signal processing approximation algorithms Speech processing Optimization

来源：评论

学校读者我要写书评

暂无评论

A 28nm 3.14 TFLOP/W BF16 LLM Fine-Tuning Processor with Asymmetric Quantization Computing for AI PC

A 28nm 3.14 TFLOP/W BF16 LLM Fine-Tuning Processor with Asym...

引用

Custom Integrated Circuits Conference (CICC)

作者： Xinyuan Lin Leran Huang Chenhan Wei Wenbin Jia Hedi Wang Wenxun Wang Hongyang Jia Huazhong Yang Yongpan Liu Tsinghua University Beijing China Tsinghua Shenzhen International Graduate School Shenzhen China

ISBN: (数字)9798331517458

ISBN: (纸本)9798331517465

The powerful capabilities of large language models (LLMs) enable them to function as personal digital assistants. To ensure user privacy, personalized fine-tuning can be conducted locally on memoryconstrained AI PCs using Parameter-Efficient Fine-Tuning (PEFT) algorithms, such as QLoRA[1] and QA-LORA[2]. Figure 1 illustrates the computation flow of QLoRA: BF16 input activations undergo matrix multiplication with 4-bit quantized pre-trained weights and BF16 adapter weights. In this flow, asymmetric quantization MACs represent the primary bottleneck, consuming approximately 97% of the computational load. However, current neural processing units (NPUs) offer limited support for asymmetric computation: fine-tuning Llama2-13B on an RTX 3090 takes over 25 hours. This highlights the need for fine-tuning processors optimized for asymmetric quantization. Yet, asymmetric quantization presents hardware design challenges: 1) Existing NPUs primarily support symmetric formats, introducing conversion overhead and inefficiencies; 2) Current NPUs lack efficient support for low-precision data transposition; and 3) 4-bit quantized QLoRA encounters high external access and storage demands, while the use of 2:4 sparsity in low-bit LLM finetuning[3] incurs substantial bitmask overhead with limited benefits.

关键词： Application specific integrated circuits Privacy Quantization (signal) Program processors Large language models Personal digital devices approximation algorithms Hardware

来源：评论

学校读者我要写书评

暂无评论

Task-Oriented Semantic Information Allocation Based on Rate Splitting for Cost Minimization

Task-Oriented Semantic Information Allocation Based on Rate ...

引用

IEEE Conference on Wireless Communications and Networking

作者： Jiarong Lu Xi Li Heli Zhang Beijing University of Posts and Telecommunications Beijing China

ISBN: (数字)9798350368369

ISBN: (纸本)9798350368376

Task-oriented semantic communication effectively facilitates the completion of specific tasks by conveying users' interests. Nevertheless, when a base station conveys task-oriented semantic information to multiple users, these users exhibit both shared and distinct interests, which leads to increased redundancy and communication costs. To reduce transmission redundancy and cost, in this paper, a flexible and efficient multicast and unicast transmission scheme based on task-oriented semantic information is investigated for users with overlapping interests. Specifically, a joint energy-delay cost function is adopted to measure the communication cost caused by transmission redundancy, which facilitates a trade-off between energy consumption and transmission delay. Then, to minimize the network cost, this paper leverages the inherent benefits of common and private streams in rate splitting multiple access (RSMA) and design a semantic information allocation strategy. The RSMA common stream is further divided into the super-common and user-common streams. A semantic information allocation mechanism is established according to the common and private streams to flexibly allocate multiple semantic information. Then, semantic information selection on super-common stream, the proportional strategy of user-common stream, and rate splitting are jointly designed to minimize the network cost. The problem is a mixed-integer problem. Block coordinate descent and successive convex approximation technique are adopted to address it. An algorithm, namely SURA, is proposed for semantic information allocation in RSMA networks. Simulation results demonstrate the proposed SURA algorithm can adaptively allocate semantic information and reduce the network cost.

关键词： Costs Multicast algorithms Unicast Simulation Semantics Redundancy Minimization approximation algorithms Resource management Streams

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：